UUID Generator (Version 4)

How it works

Type how many UUIDs you need in the left box; the page opens with 5 already generated and the list refreshes on every change. Seeding a database with 100 test records? Type 100 and you get one UUID per line — "Copy output" puts the whole list on your clipboard, and the download button saves it as a text file. The identifiers are version 4: in the 36-character form 550e8400-e29b-41d4-a716-446655440000, the first digit of the third block is always 4. Randomness comes from the browser's cryptographic source (crypto.getRandomValues) and generation happens entirely on your device — no identifier is ever sent to a server, and closing the tab destroys the list. Working with a system that expects UPPERCASE, or writing into a 32-character field that rejects dashes? Switch the format in the list — the identifier's value stays the same, only its spelling changes.

This tool is also known as uuid generator, guid generator, random uuid generator, uuid v4 generator, bulk uuid generator, generate guid online.

What is UUID?

A UUID (universally unique identifier) is a 128-bit identifier used to tell records apart; its standard spelling is 32 hexadecimal digits split by dashes, like 550e8400-e29b-41d4-a716-446655440000. Its power is needing no central registry: two computers that have never heard of each other can both generate UUIDs, and the chance of them producing the same value is practically zero. That is why UUIDs are everywhere — database keys, file and session naming, distributed systems and API records.

What is UUID version 4 (random)?

The UUID standard defines several versions, and the version number tells you how the identifier was made: version 1 builds on time and hardware address, version 5 hashes a name, and version 4 is generated entirely at random. This tool produces version 4: of the 128 bits, 6 carry version and variant information and the remaining 122 are filled from a cryptographic random source. That is why the first digit of the third block is always 4. Because random generation leaks no time or device information into the identifier, version 4 is the most widely used version today.

What is the difference between a UUID and a GUID?

In practice, none: GUID (globally unique identifier) is Microsoft's name for the same 128-bit structure, and it is the term you meet in Windows, .NET and SQL Server documentation. Both follow the same standard, use the same spelling and are fully interchangeable. The one habitual difference is case: Microsoft tooling often displays GUIDs in UPPERCASE, sometimes wrapped in curly braces, while the standard form is lowercase. The UPPERCASE option on this page exists precisely for those environments — any field asking for a GUID will accept an identifier made here.

Is the collision chance really zero?

Mathematically no, practically yes. A version 4 UUID carries 122 bits of randomness; a collision means the same 122-bit value coming up twice. Even generating a billion UUIDs per second, you would need roughly 86 years of non-stop generation before the first collision becomes a coin flip — which is why no real system bothers guarding against UUID collisions.

That guarantee rests on the quality of the randomness: a weak source (Math.random, say) can repeat values far sooner. This tool uses the browser's cryptographic source (crypto.getRandomValues). Still, a UUID is an identifier, not a secret key — for session tokens, password-reset links and anything else that must stay secret, use token generation designed for that job.

Which format for which job?

The standard form is lowercase with dashes; virtually every system that accepts UUIDs takes it as-is, and when in doubt it is the right choice. Do not let case trip up comparisons: 550E8400 and 550e8400 are the same identifier, and most systems normalize before comparing.

• Database keys, API records, general use → lowercase with dashes (standard)
• Windows registry, legacy Microsoft tools asking for a GUID → UPPERCASE with dashes
• 32-character fields that reject dashes, compact URL parts → 32 digits, no dashes
• Bulk test data → set the count to 100-500 and download the list as a file

Frequently asked questions

Can the same UUID come up twice?

In theory yes, in practice no. Each identifier carries 122 bits of randomness, and the collision probability is so small that even at a billion generations per second the expected first collision is decades away. You can use the identifiers without any duplicate checking.

Can anyone tell when or on which device a UUID was made?

Not with version 4. Version 1 UUIDs embed the creation time and the network card address inside the identifier; version 4 is entirely random apart from its version bits. Nothing about time, device or sequence can be read out of an identifier made here.

Can I use a UUID as a password or session token?

Not recommended. This tool draws from a cryptographic random source, but a UUID is an identifier format: its spelling follows a predictable pattern and many systems write UUIDs openly into logs and URLs. For secrets, use token generation designed for that purpose; for passwords, open the password generator on this site.

Is a UUID a good database primary key?

It is common and very useful in distributed systems; but because version 4 is random, rows enter the index in scattered order, which can hurt index performance under very heavy insert load. If you need ordered identifiers, time-ordered UUIDs (version 7) were designed for that — this tool generates version 4 only.